Neuropsychologia 44 (2006) 300–306

Preserved complex emotion-based learning in amnesia
Oliver H. Turnbull ∗ , Cathryn E.Y. Evans
Centre for Cognitive Neuroscience, School of Psychology, University of Wales, Bangor, Wales LL57 2AS, UK
Received 30 July 2004; received in revised form 26 April 2005; accepted 26 April 2005
Available online 18 July 2005

Abstract
An important role for emotion in decision-making has recently been highlighted by disruptions in problem solving abilities after lesion to
the frontal lobes. Such complex decision-making skills appear to be based on a class of memory ability (emotion-based learning) that may
be anatomically independent of hippocampally mediated episodic memory systems. There have long been reports of intact emotion-based
learning in amnesia, arguably dating back to the classic report of Claparede. However, all such accounts relate to relatively simple patterns
of emotional valence learning, rather than the more complex contingency patterns of emotional experience, which characterise everyday
life. A patient, SL, who had a profound anterograde amnesia following posterior cerebral artery infarction, performed a measure of complex
emotion-based learning (the Iowa Gambling Task) on three separate occasions. Despite his severe episodic memory impairment, he showed
normal levels of performance on the Gambling Task, at levels comparable or better than controls—including learning that persisted across
substantial periods of time (weeks). Thus, emotion-based learning systems appear able to encode, and sustain, more sophisticated patterns of
valence learning than have previously been reported.
© 2004 Elsevier Ltd. All rights reserved.
Keywords: Emotion; Emotion-based learning; Amnesia; Episodic memory; Decision-making

Patients with substantial lesions to the frontal lobes often
show relatively normal intelligence, and near-normal performance on a range of traditional executive measures, such as
the Wisconsin Card Sorting Task and Tower of Hanoi/London
(Bechara, Damasio, Damasio, & Anderson, 1994; Damasio,
1996). However, in spite of normal performance on ‘executive’ tasks, such patients often choose unsuitable friends,
enter inadvisable relationships and engage in ill-advised
activities (Bechara et al., 1994; Damasio, 1996; Manes et al.,
2002; Rogers et al., 1999; Rolls, Hornack, Wade, & McGrath,
1994). It has been proposed that these poor judgement and
decision-making abilities in social cognition follow from an
inability to use emotion-based knowledge about the possible outcome of decisions (Bechara, Damasio, & Damasio,
2000; Damasio, 1994, 1996; LeDoux, 2000; Rolls et al.,
1994; Turnbull, Berry, & Bowman, 2003, though see Tomb,
Hauser, Deldin, & Caramazza, 2002 for a dissenting voice).

∗ Corresponding author. Tel.: +44 1248 383670; fax: +44 1248 382599.

E-mail address: o.turnbull@bangor.ac.uk (O.H. Turnbull).
0028-3932/$ – see front matter © 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.neuropsychologia.2005.04.019

A number of different measures have been designed to
assess emotion-based learning, for example the Cambridge
Gambling Task (Rogers et al., 1999) and tasks of delayed discounting (Bickel, Degrandpre, & Higgins, 1995; Bowman &
Turnbull, 2004). However, much of the research has focused
on one measure, the Iowa Gambling Task (Bechara et al.,
1994; Bechara, Tranel, Damasio, & Damasio, 1996; Bechara,
Damasio, Tranel, & Damasio, 1997; Bechara et al., 2000).
Here, the participant is faced with four decks of cards, and
asked to choose any deck, in any sequence. They win or lose
money with each card turn. Some decks have high gains, but
also substantial losses, such that sustained playing leads to
overall financial loss. Other decks have more modest payouts, but produce smaller losses, so that sustained playing
leads to small but consistent gains. The claim that learning on
such measures is somehow related to emotion systems (rather
than other classes of non-declarative memory) is based on a
range of factors: notably that such tasks involve an explicit
regime of reward and punishment (Bechara et al., 2000) produce, modifications in skin-conductance findings while the
task is being played (e.g., Bechara, Damasio, Damasio, &

O.H. Turnbull, C.E.Y. Evans / Neuropsychologia 44 (2006) 300–306

Lee, 1999), and a series of findings demonstrating poor performance on the task after lesion to known emotion-related
brain regions such as the ventromesial frontal lobes (Bechara
et al., 1994, 1996, 2000, though see also Manes et al., 2002)
and amygdala (Bechara et al., 1999).
The Iowa group have argued that the Gambling Task is
extremely complex (Damasio, 1994), and that participants
do not appear, subjectively, to understand the contingencies of the game (Bechara et al., 1994, 1996, 1997, 2000,
though see, Bowman, Evans, & Turnbull, 2005; Evans et
al., in press; Maia & McClelland, 2004). This has implications for the sorts of memory systems that are available
for use when the game is played. Humans typically report
relying heavily on explicit episodic memory, which is mediated by hippocampal and related circuits (e.g., Squire, 1992;
Tulving & Schachter, 1990). However, Gambling Task findings have demonstrated that complex decision-making may
also rely on largely implicit emotion-based learning systems
(Bechara et al., 2000, p. 300; cf. Bowman et al., 2005; Evans
et al., in press). The neurobiology of these systems is less
well understood than those mediating episodic memory, but
they probably involve the full range of subcortical emotion
systems (Berridge, 2003; Cadler, Lawrence, & Young, 2001;
Davidson & Irwin, 1999; LeDoux, 2000; Panksepp, 1986,
1998; Patterson & Schmidt, 2003; Phan, Wager, Taylor, &
Liberzon, 2002; Rolls, 2000), as well as the connections
between these systems and pre-frontal cortex, through the
ventromesial frontal lobes (Bechara, 2004; Bechara et al.,
2000; Davidson & Irwin, 1999).
It is of especial note that the explicit episodic memory, and
emotion-based learning, systems appear to be anatomically
independent (Tranel & Damasio, 1993). In principle, therefore, it should be possible to disrupt episodic memory and yet
leave emotion-based learning intact. In support of this claim,
there have long been reports of intact emotion-based learning in people with amnesia (for review see Eichenbaum &
Cohen, 2001). The classic case is that reported by Claparede
(1911), who concealed a pin in his palm before shaking
the hand of an amnesic patient. On the day following this
painful episode, the patient refused to shake the physicians
hand, despite having no conscious recollection of the incident
(Claparede, 1951). Recent investigations of this phenomenon
have been more detailed (e.g., Damasio, Tranel, & Damasio,
1989; Johnson, Kim, & Risse, 1985; Tranel & Damasio, 1990,
1993). For example, Damasio and co-workers described the
profoundly amnesic patient, Boswell, who seemed to consistently avoid, or favour, certain hospital staff, though he
was unable to explicitly remember the relevant individuals
(Tranel & Damasio, 1990, 1993). In a more formal investigation, Boswell was engaged in three types of inter-personal
encounter, with individuals who played a ‘good’, ‘neutral’
or ‘bad’ character in all their interactions. At the end of the
week Boswell was shown sets of photographs that included
the face of one of the individuals, and an unfamiliar face,
and was asked to “Pick the person you would like best?”
(p. 83). Although he had no explicit memory for any of

301

theindividuals, Boswell chose the ‘good’ character some
80% of the time, and almost never chose the bad character
(Tranel & Damasio, 1993). Investigations of this type (see
also Johnson et al., 1985) suggest that profoundly amnesic
patients may retain a capacity to learn emotional valence,
demonstrating a single dissociation between emotion-based
learning and declarative memory.
However, reports of the sort described above relate to
relatively simple patterns of emotional valence learning.
For example, in the experimental design of the Tranel and
Damasio (1993) study, Boswell encountered individuals who
were ‘good’ in all of their inter-personal encounters. He was
not exposed to the more complex patterns of valence, which
characterise everyday life (e.g., Barraclough, Conroy, & Lee,
2004). Here individuals often behave in ways that might be
perceived as positive and negative on different occasions,
though such individuals often can still be judged as ‘good’ and
‘bad’ based on some overall, aggregate, assessment of their
complex contingency history. It is precisely this complicated
pattern of reward and punishment that the Iowa Gambling
Task was designed to assess (Bechara et al., 2000).
Of course, this begs the question of what constitutes ‘complexity’ on tasks of this type. For example, the Wisconsin
Card Sorting Task (WCST) is an apparently complex card
game, which provides some degree of reward and punishment in its feedback to the patient, through a verbal response
(right or wrong) by the examiner. Apparently normal performance in the WCST in amnesia has been reported (e.g.,
Leng & Parkin, 1988; Moudgil, Azzouz, Al-Azzaz, Haut, &
Gutmann, 2000; Shoqeirat, Mayes, MacDonald, Meudell, &
Pickering, 1990, though see also Squires, Hunkin, & Parkin,
1996; Stefanacci, Buffalo, Schmolck, & Squire, 2000). Similarly, preserved learning in amnesia has been shown on the
‘Weather Prediction’ task (WPT) that involves the probabilistic predictive power of a series of four cards (Knowlton,
Squire, & Gluck, 1994; Knowlton, Mangels, & Squire, 1996).
The Iowa Gambling Task (IGT) differs from these measures,
as regards apparent ‘complexity’, in several respects. Firstly,
there is the question of the number of ‘objects’ to be tracked.
The IGT involves the tracking of four independent decks of
cards, unlike the single deck of the WCST. Also, while the
WPT does involve four independent cards, very high performance levels can be achieved by focussing only on a
single deck, the most commonly employed strategy (Gluck,
Shohamy, & Myers, 2002). Secondly, the IGT repeatedly
involves ambiguity of valence: such that a single deck can
produce a series of events of randomly-mixed reward and
punishment, while the WCST produces consistent reward
or punishment when the participant retains a single sorting
strategy. Finally, the IGT involves a lengthy series of trials
(100), which is far longer than the modified-WCST (48). The
WPT has been administered for periods over 100 trials (e.g.,
Gluck et al., 2002) but is notable that the original Knowlton
et al.’s (1994, 1996) studies showed that amnesic participants
were out-performed by controls after 50 trials, as declarative
knowledge became more advantageous for non-amnesics.

302

O.H. Turnbull, C.E.Y. Evans / Neuropsychologia 44 (2006) 300–306

Thus, it would be of some interest to observe whether
normal Iowa Gambling Task performance can be maintained
even in the presence of profound amnesia. The present study
demonstrates that it is possible to show normal levels of memory ability on a complex emotion-based learning task, in
the presence of profound episodic memory impairment. A
patient, SL, who had a dense anterograde amnesia following
posterior cerebral artery infarction, was assessed. He performed the Iowa Gambling Task on three occasions, over 3
weeks; and also completed a range of measures of memory
and executive abilities. On the Gambling Task he showed normal levels of performance, comparable or better than controls
– showing intact complex emotion-based learning – including
learning which persisted across substantial periods of time.

1. Case study
1.1. History
SL was an 85-year-old married man, with two grown
up children, who worked as a building inspector before he
retired. He suffered a posterior cerebral artery stroke in late
2003, 2 months prior to the initial assessment. CT scans indicated a large hypodense area across almost all of the left
posterior cerebral artery territory, consistent with a recent
infarct (see Fig. 1). On the left side, the lesion involved
the medial and lateral surfaces of the occipital lobe, as well
as the infero-medial regions of the posterior temporal lobe,
including the posterior regions of the hippocampus and the
para-hippocampal gyrus. It is not clear whether the damage

involved regions of the right posterior cerebral artery territory that were not visible on scan. However, SL presented
with a profound amnesia, and also reported a colour recognition impairment (achromatopsia) in both visual fields, which
is consistent with the possibility that both posterior cerebral
artery territories were involved.
In spite of his profound amnesia, SL’s behaviour towards
the investigators appeared to gradually change across sessions (cf. Tranel & Damasio, 1993, p. 80). In the earlier
sessions, SL appeared to be relatively uncertain about the
investigators, and would frequently ask questions about the
rationale of his assessment: such as “Who are you?”, “Where
did you come from?”, and “What am I doing?”. SL never
explicitly recalled who the investigators were, but the uncertainty about people and setting greatly lessened across assessment sessions. Despite never recognising the investigators or
the setting in later sessions, he was far more relaxed during
assessments, and the repetitive questions about identity and
rationale were no longer present.
1.2. Memory
SL displayed marked disorientation for time and place.
He knew his name and date of birth, but described his age as
‘mid-sixties’. He reported the date as “probably some time in
the early 1980’s”. He was unable to name the current Prime
Minister, and guessed it was probably Margaret Thatcher.
When asked the age of his children (actually in their forties)
SL claimed they were teenagers. He did not recognise any of
the regular nursing staff, and could not recall his wife’s visits.
Even after several visits associated with his neuropsychological assessment, he had no recollection of the examiners, or
the different tasks he had performed. Indeed, he did not recall
the examiner even when they left the room for a few minutes.
SL’s memory and learning abilities were assessed using
the Wechsler Memory Scale III (Wechsler, 1997):
1.3. Measures of recent episodic memory

Fig. 1. CT scan for SL, showing posterior cerebral artery infarction.

Information: 3/18. SL was able to report only his name, date
of birth, and mother’s first name
Logical Memory: 0/50
Logical Memory Delayed: 0/50
Logical Memory Delayed Recognition: 17/30 (where a
chance score is 15/30)
Face Recognition: 23/48 (where a chance score is 24/48)
Face Recognition Delayed: 20/48 (where a chance score is
24/48)
Verbal paired Associates: 0/32
Verbal paired Associates Delayed: 0/8
Verbal paired Associates Delayed Recognition: 20/24
(where a chance score is 12/24)
Family Pictures: 0/64
Family Pictures Delayed: 0/64
Visual Reproduction: 0/104

O.H. Turnbull, C.E.Y. Evans / Neuropsychologia 44 (2006) 300–306

1.4. Measures of immediate/working memory
Mental Control: 16/16
Digit Span: SL had an audio-verbal span of 7 items forwards,
and 2 backwards
Letter-number sequencing: SL had a span of 3 items
Spatial Span: SL had a spatial span of 4 items forwards, and
4 backwards
1.5. Executive functions
Assessment of executive function was limited by SL’s
achromatopsia (e.g., he could not be tested on the Wisconsin Card Sorting Task, or the Tower of London). However,
he did not appear to be impaired on the tests of executive
function that he was able to complete (Benton & Hamsher,
1989; Shallice & Evans, 1978; Wilson, Alderman, Burgess,
Emslie, & Evans, 1996).
Controlled Oral Word Association Test: total score 31, average
Cognitive Estimates Test: 15/15
Behavioural Assessment of the Dysexecutive Syndrome Rule
Shift sub-test: 4/4, average
1.5.1. Comment
SL’s performance suggests severe impairments in laying
down recent episodic memories, regardless of presentation
format. In contrast, he had an immediate memory span within
the normal range on a number of tasks, especially on simple
measures of audio-verbal recall, such as digit span forwards.
One remarkable finding was a surprisingly good performance, well above chance (20/24), in his delayed recognition
of paired associate items—despite a failure to bring even a
single one of the pairs to conscious recall on any of 40 previous exposures to the pairs.
1.6. Iowa Gambling Task
SL completed a version of the Iowa Gambling Task identical to the original Bechara et al. (1994) study. He could
choose any card from the four decks (labelled A, B, C, D) in
any sequence. Decks A and B were disadvantageous, and C
and D advantageous, though the schedule of reward and punishment differed. In Deck A (bad deck) and Deck C (good
deck) there were five smaller unpredicted punishments per 10
card selections (thus 20 punishments in total for each deck).
On Decks B and D there was only 1 punishment per 10 card
selections, which was equal in size to the total 10 card loss on
Decks A and C. The only minor difference in administration
from that of Bechara et al. was the value of the reward and
punishment (see Bowman & Turnbull, 2003). SL was given
£2.00 worth of real money, where US$ 1000 on the Bechara
et al. task was equal to £1 in the present study. For the disadvantageous decks (A and B), he won 10p for every card turn,
incurring losses of between 15p and £1.25. On the advanta-

303

geous decks (C and D), SL won 5p for every card turned,
incurring losses of between 2p and 25p. He was informed at
the start of the game, and regularly throughout the game that
he could keep any money they won.
As in Bechara et al. (1994), the task continued until SL
had drawn 100 cards. However, it was necessary to explain
rules of the task on several occasions, when he asked what
task he was performing. SL completed the Gambling Task on
three separate sessions, each spaced 1 week apart. In order
to counteract any simple effects of learning, such as the fact
that Decks C and D were the ‘good’ decks in the first administration, the contingencies were shifted between sessions. In
Session 1, Decks C and D were ‘good’; in Session 2, Decks
A and D were ‘good’; and in Session 3, Decks A and B were
‘good’ (for more details on the rationale for this shifting pattern see Turnbull et al., in press). When first exposed to a task
in a given week, SL never recalled performing the task in a
previous session.
Four gender and age matched controls (mean age: 81)
were recruited through the Bangor University Community
Participant Panel. These participants were also tested on the
Gambling Task three times, at 1 week intervals. The controls
had no history of neurological or psychiatric disorder.

2. Results
Data were analysed in two separate ways: firstly, to
observe whether there was learning during the individual
sessions, and secondly whether there were any cumulative
effects of learning across sessions.
2.1. Learning during individual sessions
As in Bechara et al. (1994), the 100 card selections were
sub-divided into five blocks of 20 trials each. The net score
of each block was calculated by subtracting the number of
bad from good card selections [(C + D) − (A + B)]. A net
score above zero implied that the participants were selecting
cards advantageously. For the purposes of this investigation,
the control participants and SL’s performance was averaged
across each of the three sessions.
There was a progressive increase, across blocks, in advantageous card selections for both SL and the control participants (see Fig. 2). Performance levels began close to chance,
and increased to levels substantially above chance by Block 5.
Single case study t-tests (see Crawford & Garthwaite, 2002,
for further detail) were employed to investigate a possible
difference between SL and the controls performance on the
Gambling Task (averaged across the three sessions). There
were no significant differences between SL and controls’ performance in four of the five blocks (t-values ranging from 0.20
to 0.53, all significance levels p > 0.05). SL’s performance
was significantly better than controls in block 2 (t = 3.71,
p < 0.05).

304

O.H. Turnbull, C.E.Y. Evans / Neuropsychologia 44 (2006) 300–306

Fig. 2. Performance of SL and controls on the Iowa Gambling Task, averaged
across three sessions.

Fig. 3. Performance of SL and controls on the Iowa Gambling Task, averaged
across block, over three sessions.

2.2. Cumulative effects of learning across sessions
To investigate performance over the three sessions, a net
score for the entire task was calculated by subtracting the
number of good from bad card selections [(C + D) − (A + B)]
for all 100 trials. Again, a net score above zero implied that
the participants were selecting cards advantageously.
There was a progressive increase in advantageous card
selections across sessions (see Fig. 3) for both SL and the controls. Again, t-tests revealed no significant difference between
SL and controls’ performance in any session (Session 1
(t = 0.06, p = 0.48); Session 2 (t = −0.12, p = 0.46); Session
3 (t = 0.83, p = 0.24)). Notably, however, SL performed (nonsignificantly) better than controls in Session 3.

3. Discussion
The key aim of the present study was to evaluate the
performance of an amnesic patient on a measure of complex emotion-based learning. SL presented with a profound
impairment of recent episodic memory, as shown by a
range of floor, or chance, performances. In contrast, he
showed reasonable levels of performance on tasks of immediate/working memory. This pattern of performance is entirely
consistent with the sorts of episodic amnesia typically

observed after lesion to the hippocampal memory system
(Squire, 1992).
However, in striking contrast to his poor performance on
tasks of recent episodic memory, SL showed apparently normal levels of performance on the Iowa Gambling Task, even
out-performing age-matched controls on some blocks. This
learning took place despite SL being unable to recall the rules
of the task for more than a few consecutive trials, and being
unable to recall anything of the task directly after performing it. While preserved learning on card game tasks has been
reported before in amnesia (e.g., Knowlton et al., 1994, 1996;
Leng & Parkin, 1988; Moudgil et al., 2000; Shoqeirat et al.,
1990), the preserved learning in SL is reported on a task
that is (as mentioned in the Introduction) notably more complex than those previously employed: in terms of number of
independent decks, number of trials, and most importantly
because of ambiguity of valence.
This latter finding, of entirely normal levels of performance on a 100 trial task, is especially striking given that the
Knowlton et al. ‘Weather Prediction’ studies (1994, 1996)
showed that amnesic participants were eventually outperformed by controls after 50 trials, presumably because declarative knowledge became more advantageous. However, the
Iowa Gambling Task has a more explicit regime of reward
and punishment (winning and losing real money) than the
‘Weather Prediction’ task, and it may be that an amnesic
such as SL is better able to capitalise on more powerfully
experienced emotion-based knowledge. Indeed, SL appeared
to perform at even higher levels than controls in the later
blocks, after more than 200 trials. This may be because controls attempt to solve the Iowa Gambling Task by using
explicit conceptual strategies, not available to SL, though
such approaches may be counter-productive (Evans, Kemish,
& Turnbull, 2004).
SL was also able to demonstrate a regular and systematic
improvement in performance over three test sessions, each
separated by 1 week: where his performance was comparable, or better, than controls. This improvement occurred
despite the fact that the reward-contingency pattern was
shifted between sessions, and that SL was unable to explicitly
recall any aspect of his performance in the previous sessions,
or even to recognise the examiner. This ability to retain and
build on established emotion-based knowledge, across long
periods of time, suggests that these memory systems have
substantial levels of temporal stability, with some additional
capacity for flexibility.
One especially interesting finding was a remarkably good
performance, well above chance levels, in SL’s recognition
of paired-associate items. He had comprehensively failed to
bring even a single one of these pairs to conscious recall
on any his 40 previous exposures to the pairs, but nevertheless appeared to have encoded at least some aspect of a
memorial linkage between them. One possibility is that he
had stored some emotional marker associated with each pair
(‘rose–bag’, good; ‘elephant–glass’, bad). Another possibility is that the previously tested items had acquired some pos-

O.H. Turnbull, C.E.Y. Evans / Neuropsychologia 44 (2006) 300–306

itive emotional valence through the ‘mere-exposure’ effect
(Zajonc, 1980). These arguments remain speculative, but
clearly this finding bears on the important question of
how episodic memory abilities are bolstered by additional
resources from other memory systems, and this issue merits replication and further investigation in other amnesic
patients.
These findings support the consensus that there are multiple memory systems in the brain, each with functionally and
anatomically distinct neural substrates, especially supporting
an anatomical and functional dissociation between episodic
and emotion-based memory. Importantly, the emotion-based
learning systems appear able to encode more sophisticated
patterns of valence learning than have previously been
reported, and sustain these over substantial periods of time.
However, to further investigate this topic, it will be of some
interest to assess a range of amnesic individuals, for example, establishing whether such learning is universally found
in amnesic patients, and whether the effect can be found after
amnesia of the ‘diencephalic’ sub-type, in addition to those
with ‘hippocampal’ amnesia.

Acknowledgement
We would like to thank Prof. Robert Rafal for his assistance with the neurological and neuroradiological aspects of
the paper.
This research was funded by the National Alliance for
Research on Schizophrenia and Depression (NARSAD) and
the Wellcome Trust.

References
Barraclough, D. J., Conroy, M. L., & Lee, D. (2004). Prefrontal cortex
and decision-making in a mixed-strategy game. Nature Neuroscience,
7, 404–410.
Bechara, A. (2004). A neural view of the regulation of complex cognitive
functions by emotion. In P. Philippot & R. S. Feldman (Eds.), The
regulation of emotion. Mahwah, NJ: Lawrence Erlbaum Associates.
Bechara, A., Damasio, H., & Damasio, A. R. (2000). Emotion, decisionmaking and the orbitofrontal cortex. Cerebral Cortex, 10, 295–307.
Bechara, A., Damasio, A. R., Damasio, H., & Anderson, S. W. (1994).
Insensitivity to future consequences following damage to prefrontal
cortex. Cognition, 50, 7–15.
Bechara, A., Damasio, H., Damasio, A. R., & Lee, G. P. (1999). Different contributions of the human amygdala and ventromedial prefrontal
cortex to decision-making. The Journal of Neuroscience, 19(13),
5473–5481.
Bechara, A., Damasio, H., Tranel, D., & Damasio, A. R. (1997). Deciding
advantageously before knowing the advantageous strategy. Science,
275, 1293–1294.
Bechara, A., Tranel, D., Damasio, H., & Damasio, A. R. (1996). Failure
to respond autonomically to anticipated future outcomes following
damage to prefrontal cortex. Cerebral Cortex, 6, 215–225.
Benton, A. L., & Hamsher, K. (1989). Multilingual aphasia examination.
Iowa City, IA: AJA Associates.
Berridge, K. C. (2003). Pleasures of the brain. Brain and Cognition, 52,
106–128.

305

Bickel, W. K., Degrandpre, R. J., & Higgins, S. Y. (1995). The behavioural economics of concurrent drug reinforcers—A review and
reanalysis of drug self-administration research. Psychopharmacology,
118, 250–259.
Bowman, C. H., & Turnbull, O. H. (2003). Real versus facsimile reinforcers on the Iowa Gambling Task. Brain & Cognition, 53, 207–210.
Bowman, C. H., & Turnbull, O. H. (2004). Emotion-based learning on
a simplified card task: The Iowa and Bangor Gambling Tasks. Brain
and Cognition, 55, 277–282.
Bowman, C. H., Evans, C. E. Y., & Turnbull, O. H. (2005). Artificial time-constraints on the Iowa Gambling Task: The effects on
behavioural performance and subjective experience. Brain and Cognition, 57, 21–25.
Cadler, A. J., Lawrence, A. D., & Young, A. W. (2001). Neuropsychology
of fear and loathing. Nature Reviews, 2, 352–363.
Claparede, E. (1951). Recognition and “me”ness. In D. Rapaport (Ed.),
Organisation and pathology of thought (pp. 58–75). New York:
Columbia University Press (Reprinted from Archives de Psychologies,
11, 79–90 (1911)).
Crawford, J. R., & Garthwaite, P. H. (2002). Investigation of the single
case in neuropsychology: Confidence limits on the abnormality of test
scores and test score differences. Neuropsychologia, 40, 1196–1208.
Damasio, A. R. (1994). Descartes’ error: Emotion, reason, and the human
brain. New York: Putnam.
Damasio, A. R. (1996). . The somatic marker hypothesis and the possible
functions of the prefrontal cortex: Vol. B351. London: Philosophical
Transactions of the Royal Society, pp. 1413–1420.
Damasio, A. R., Tranel, D., & Damasio, H. (1989). Amnesia caused by
herpes simplex encephalitis, infarctions in basal forebrain, Alzheimer’s
disease, and anoxia. In F. Boller & J. Grafman (Eds.), Handbook of neuropsychology: Vol. 3. Amsterdam: Elsevier, pp. 149–
166.
Davidson, R. J., & Irwin, W. (1999). The functional neuroanatomy of
emotion and affective style. Trends in Cognitive Sciences, 3, 11–21.
Eichenbaum, H., & Cohen, N. J. (2001). From conditioning to conscious
recollection: Memory systems of the brain. Oxford: Oxford University
Press.
Evans, C. E. Y., Bowman, C. H., & Turnbull, O. H. Subjective awareness
on the Iowa Gambling Task: The key role of emotional experience in
schizophrenia. Journal of Clinical and Experimental Neuropsychology,
in press.
Evans, C. E. Y., Kemish, K., & Turnbull, O. H. (2004). Paradoxical
effects of education on the Iowa Gambling Task. Brain & Cognition,
54, 240–244.
Gluck, M. A., Shohamy, D., & Myers, C. (2002). How do people solve
the ‘Weather Prediction’ task? Individual variability in strategies for
probalistic category learning. Learning & Memory, 9, 408–418.
Johnson, M. H., Kim, J. K., & Risse, G. (1985). Do alcoholic Korsakoff’s
syndrome patients acquire affective reactions? Journal of Experimental Psychology: Learning Memory and Cognition, 11, 27–36.
Knowlton, B., Squire, L. R., & Gluck, M. (1994). Probabilistic classification learning in amnesia. Learning and Memory, 1, 106–
120.
Knowlton, B., Mangels, J., & Squire, L. R. (1996). A neostriatal habit
learning system in humans. Science, 273, 1399–1402.
LeDoux, J. E. (2000). Emotion circuits in the brain. Annual Review of
Neuroscience, 23, 155–184.
Leng, NR., & Parkin, A. J. (1988). Double dissociation of frontal dysfunction in organic amnesia. British Journal of Clinical Psychology,
27, 359–62.
Manes, F., Sahakian, B., Clark, L., Rogers, R., Antonin, N., Aitken, M.,
et al. (2002). Decision-making processes following damage to the
prefrontal cortex. Brain, 125, 624–639.
Maia, T. V., & McClelland, J. L. (2004). A re-examination of the evidence
for the somatic marker hypothesis: What participants really know in
the Iowa Gambling task. Proceedings of the National Academy of
Sciences, 101, 16075–16080.

306

O.H. Turnbull, C.E.Y. Evans / Neuropsychologia 44 (2006) 300–306

Moudgil, S. S., Azzouz, M., Al-Azzaz, A., Haut, M., & Gutmann, L.
(2000). Amnesia due to fornix infarction. Stroke, 31, 1418.
Panksepp, J. (1986). The anatomy of emotions. In R. Plutchik & H.
Kellerman (Eds.), Biological foundations of emotions: Vol. 3. Emotion,
theory, research and experience. Academic Press.
Panksepp, J. (1998). Affective neuroscience: The foundations of human
emotions. Oxford: Oxford University Press.
Patterson, D. W., & Schmidt, L. A. (2003). Neuroanatomy of the human
affective system. Brain and Cognition, 52, 24–26.
Phan, K. L., Wager, T., Taylor, S. F., & Liberzon, I. (2002). Functional
neuroanatomy of emotion: A meta-analysis of emotion activation studies in PET and fMRI. Neuroimage, 16, 331–348.
Rogers, R. D., Owen, A. M., Middleton, H. C., Williams, E. J., Pickard,
J. D., Sahakian, B. J., et al. (1999). Choosing between small, likely
rewards and large, unlikely rewards activates inferior and orbital prefrontal cortex. The Journal of Neuroscience, 20(19), 9029–9038.
Rolls, E. T. (2000). Precis of the brain and emotion. Behavioural and
Brain Sciences, 23, 177–234.
Rolls, E. T., Hornack, J., Wade, D., & McGrath, J. (1994). Emotion
related learning in patients with social and emotional changes associated with frontal lobe damage. Journal of Neurology, Neurosurgery
and Psychiatry, 57, 1518–1524.
Shallice, T., & Evans, M. E. (1978). The involvement of the frontal lobes
in cognitive estimation. Cortex, 14, 294–303.
Shoqeirat, M. A., Mayes, A., MacDonald, C., Meudell, P., & Pickering, A. (1990). Performance on tests sensitive to frontal lobe lesions
by patients with organic amnesia: Leng & Parkin revisited. British
Journal of Clinical Psychology, 29, 401–408.
Squire, L. R. (1992). Memory and the hippocampus: A synthesis from
findings with rats, monkeys, and humans. Psychological Review, 99,
195–231.

Squires, E. J., Hunkin, N. M., & Parkin, A. J. (1996). Memory notebook
training in a case of severe amnesia. Neuropsychological Rehabilitation, 6, 55–66.
Stefanacci, L., Buffalo, E. A., Schmolck, H., & Squire, L. R. (2000).
Profound amnesia after damage to the medial temporal lobe: A
neuroanatomical and neuropsychological profile of patient E.P. The
Journal of Neuroscience, 20, 7024–7036.
Tomb, I., Hauser, M., Deldin, P., & Caramazza, A. (2002). Do somatic
markers mediate decisions on the gambling task? Nature Neuroscience, 14, 1099–1108.
Turnbull, O. H., Berry, H., & Bowman, C. H. (2003). Direct versus
indirect emotional consequences on the Iowa Gambling Task. Brain
and Cognition, 53, 389–392.
Turnbull O. H., Evans C. E. Y., Kemish K., Park S., Bowman C. H. (in
press). A novel ‘set-shifting’ modification of the Iowa Gambling Task:
Flexible emotion-based learning in schizophrenia. Neuropsychology.
Tranel, D., & Damasio, A. R. (1990). Covert learning of emotional
valence in patient Boswell. Journal of Clinical Psychology and Experimental Neuropsychology, 12, 27.
Tranel, D., & Damasio, A. R. (1993). The covert learning of affective
valence does not require structures in hippocampal system or amygdala. Journal of Cognitive Neuroscience, 5(1), 79–88.
Tulving, E., & Schachter, D. (1990). Priming and human memory systems.
Science, 247, 301–306.
Wilson, B. A., Alderman, N., Burgess, P., Emslie, H., & Evans, J. J.
(1996). Behavioural assessment of the dysexecutive syndrome. Thames
Valley Test Company.
Wechsler, D. (1997). Wechsler Memory Scale (3rd ed.). San Antonio, TX:
The Psychological Corporation.
Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. American Psychologist, 35, 547–563.